Why there's no "at" command?

[neurino]

Why happens this?

Code: Select all

# at
bash: at: command not found

[Lobster]

Puppy uses busybox
http://www.busybox.net/about.html

[Bruce B]

After two years, you know Puppy is a mini-distro. If it had all utilities and
features, it wouldn't be a mini-distro.

Also, I think you are the first person to request it.

Happy to share it with you.

~

[neurino]

Thanks Bruce,

I know Puppy is a mini-distro (and know Busybox too, Lobster, thanks) and agree with your point of view but only yesterday I knew about tac, that's available in Puppy, so I wondered why not at too which, in my humble opinion, is a bit more notorious and used...

Thanks for sharing

P.S.: I looked for at source to compile by myself, like I did with less times ago (busybox version lacks of options needed by ipython), but could not find it... where does your binary come from?

[miriam]

Actually, I'm someone who loves the fact that Puppy is a minimal distro. One of the cool things is that it doesn't start bloated, but lets you add capabilities as you need them... except sometimes...

'at' is one such function that I'd love to have. The 'at' command on its own is useless without the daemon, 'atd'. I've tried installing it from debian binaries, but still can't get the daemon running, so of course it is pointless running 'at'. Looks like I'll have to compile it all... at least then it might let me know what dependencies it requires.

[miriam]

Okay, solved it!

Get the Slackware 'at' binary package:
http://mirrors.slackware.com/slackware/ ... i486-1.txz
or if links have changed go to
http://www.slackware.com/
then take the link to where their packages are and search for 'at' (without the quotes of course). It will return a list of packages, one of which will be 'at'.

In recent Puppies you can install the txz archive simply by clicking on it after you've downloaded it. It will ask you want to install or just uncompress it.

After it is installed you'll need to run the at daemon, 'atd'.

Code: Select all

atd

Now you can run 'at' on the command line. I'm still learning about it but here is an example I've found works:

Code: Select all

echo "aplay /usr/share/audio/2barks.au" >~/at.list
at 1:25 pm <~/at.list

or you can give the whole command immediately like this:

Code: Select all

at 1:31 pm <<<"aplay /usr/share/audio/2barks.au"

Don't forget the quotes around the command.

You can check that the command is waiting in the queue by typing

Code: Select all

atq

and it will tell you that there is a job waiting to be run, and will give the time that it's supposed to run.

When it installs it just installs man files for documentation, but I prefer html files so I converted them and have uploaded them to my site at:
http://miriam-english.org/files/at-docs/

Now I need to read the docs and a lot more to learn all the ins and outs of this versatile command.

[tallboy]

Hi miriam.
I suggest you try THE source, specifically http://www.tldp.org/LDP/GNU-Linux-Tools ... index.html or http://www.tldp.org/LDP/abs/html/timedate.html#ATREF.

Some of the guides may seem old, but they are the basic building blocks, so they are still valid, lots of useful reading there...

tallboy

[miriam]

I don't go back and revisit The Linux Documentation Project (ldp) often enough. There is a lot more there than last time I visited. My internet connection is very very slow so I tend to download a lot of ebooks and refer to them offline (I have a large documentation directory of about 30GB on my machine, which is a lot of stuff when you're talking about text and html -- more than 200,000 files).

The Advanced Bash Scripting Guide (abs) is a wonderful resource, and I refer to it often. Everybody should keep a copy on their machine. It does get updated from time to time, and my local copy was version 5.1.07. The latest one at ldp is version 6.5 (updated earlier this year), so thanks to your prompting I now have the current one. Yay!

And I'm grateful for the pointer to the Gnu-Linux Tools Summary. Its latest date is 2006, so I don't know how I missed it before. It is pretty basic, but still has lots of neat info -- many fun-filled hours of reading there.

[tallboy]

Good to meet a fellow TLDP reader, Miriam. Too many people are unaware of all the free documentation available, and therfore also unaware of all the cool builtin Linux/Unix tools.

I was also without a personal internet connection, and had to download to a memory stick to feed my hungry Debian at home, but I never got past 4.1 Gb in /usr/share/doc/! Adding up all other manuals and info placed in different folders, I end up with close to 7.8 Gb in total. If starting from scratch, it takes a lifetime to read it all!

BTW, unless you already use it, Pinfo is a Lynx-like reader for info-files, works with the arrow-keys for lightning quick browsing/reading!

tallboy

[miriam]

heheh You're right. It would take a lifetime (or more) to read it all. I have even more text in my fiction folder (I'm a frequent visitor of Project Gutenberg and other ebook sites). The non-fiction mostly gets searched as reference. I figure one day I'll lose my net connection or be laid up sick in bed and unable to get onto the net. I have a gazillion projects I'm always trying to do, but if I just wanted to veg out I have years of reading, fiction and non-fiction, on all sorts of wonderful subjects.

I'd never heard of pinfo before. Info files tend to be forgotten too much these days, even by information obsessives like me. A while back I installed info in my Puppy in order to read them. They are a little frustrating though, now we have been so spoiled by the much easier hypertext of the web. I tried pinfo after you suggested it, but unfortunately it exits after being unable to find nogroup. That was fixed by using the addgroup command to add the nogroup group (which is weird, like pondering whether the set of empty sets contains itself). Anyway, pinfo then failed with the error message "Couldn't open temporary file".

Then I downloaded it from sourceforge and compiled it, but when it ran it gave the error message

Code: Select all

Przemek's Info Viewer v0.6.8
sh: /tmp/fileF08qXG: Permission denied
sh: /tmp/filesk9E53: Permission denied
sh: /tmp/filesk9E53: Permission denied
Error: could not open info file, trying manual
Error: No manual page found

I guess I'll have to continue to use the old info command for a while yet.

A while back I tried to find a program to convert info files into the more universal html format, without much luck. I guess I should try writing a converter someday as there does seem to be a lot of information locked up in that unfortunately obsolete format.

Coming back to the topic of at, I found a nice program called when, which seems more versatile than at or cron. It is downloadable from http://www.lightandmatter.com/ where you'll also find a number of very cool reference books on physics and calculus. I don't bother with dead-end, paper-targetted formats like pdf; I use wget's spider capabilities to grab the online html versions of the books -- much easier to use. For example,

Code: Select all

wget -m -np -k -p -E -l 0 -R.pdf,.gz,.tar,.tgz,.ps,.exe,.bz2,.sig,.zip,.rpm,.deb,.dmg,.mpg,.mpeg,.mp2,.mp3,.mp4,.avi,.mov,.qt,.swf,.wav,.au,.iso http://www.lightandmatter.com/html_books/lm

the "-R" followed by the long list of extensions prevents it wasting time downloading those format files. it will decend to deeper levels, but not back up to parent levels, will grab all the images required for the pages (even if they're on other sites) and will convert all links to work locally. Ummm... I seem to have come back to books again.

[GustavoYz]

You can do something like this too:

Code: Select all

( (sleep 10m; someCommand) & )

where "m" equals "minutes", "h" hours...

[miriam]

Yep, you're right. I use something which has slowly grown from exactly that, GustavoYz.

I began with simply typing something like, for example sleep 1h ; wavplay ding.wav but after a while this got tedious and gradually it grew through various incarnations til now I have a roxapps program I wrote whose icon sits on my desktop. When I click on it, it opens a dialog box where I can enter a message, a delay, and units (it remembers the previous items used). When I hit OK it opens a progress dialog which counts down. When the time is up it rings a bell 5 times and displays my message in a dialog box.

If anybody is interested in it I guess I can make it into a pet, though I'd need to standardise it a bit. For example it uses gtkdialog for the data entry and progress bar, but zenity for the final notification. (Zenity is extremely easy to use and I was feeling lazy.)

I've organised my own system for sounds where various sound files are kept in folders inside a custom folder /usr/share/audio/system_sounds/. I wrote a simple script that plays these sounds as named by the folders they're in, so that I could easily change the "alert" sound by dropping a different file in the "alert" folder, or the "done" sound by putting a different file in the "done' folder, and so on. This also let me put more than one sound file in a folder and it would play them each. I will eventually set up parameter detection too so that it can play a sound file at a defined rate or with echo or flanging, etc., but I haven't added that yet. Another thing I intend, but haven't done yet is to set up parameter files that use sox's play command to synthesise sounds on-the-fly -- an amazing, little-known capability. For example try this in a terminal (you need sox):

Code: Select all

play -n synth -j 3 sin %3 sin %-2 sin %-5 sin %-9 sin %-14 sin %-21 fade h .01 2 1.5 delay 1.3 1 .76 .54 .27 remix - fade h 0 2.7 2.5 norm -1

...or this:

Code: Select all

play -n synth pl G2 pl B2 pl D3 pl G3 pl D4 pl G4 delay 0 .05 .1 .15 .2 .25 remix - fade 0 4 .1 norm -1

But what I wanted more than anything is to give the command a target time instead of a sleep delay. And then I found the at command. Yay! Now I just need to set aside some time to utilise that in my remind program.